### Impact of estimate quality on risk assessments

Posted:

**Fri Jan 11, 2019 2:28 pm**How does estimate quality affect the quality of a risk assessment?

This forum includes discussions about project risk analysis software and project risk management software RiskyProject: theory, software features, best practices, possible problems, and user experience.

http://intaver.com/IntaverFrm/

Page **1** of **1**

Posted: **Fri Jan 11, 2019 2:28 pm**

How does estimate quality affect the quality of a risk assessment?

Posted: **Fri Jan 11, 2019 2:32 pm**

In of themselves, the quality or accuracy of your cost and schedule estimates are a major risk a does not matter whether you are performing Monte Carlo or Critical Path analysis. The quality of any analysis is dependent upon the quality or accuracy of the inputs or as more commonly known

Garbage In and Garbage Out (GIGO). It is the recognition that any single point estimate for schedule or cost most likely wrong that Monte Carlo risk analysis was developed. The statistical distributions with the 3 pt estimates are an acknowledgement that all estimates have a degree of uncertainty and the range is a model of how much or how little confidence people have in the original single point estimate.

In the example below, we can see how you might model a 10-day activity when you have either very high or very low confidence in the quality of the estimate. While they both have a 10 day base duration, the range between the low and the high estimates are a function of the quality of the estimate. In the image, the white bar represents the base duration (10 days) the blue bar represents the possible range of outcomes based on the quality of the estimate.

Garbage In and Garbage Out (GIGO). It is the recognition that any single point estimate for schedule or cost most likely wrong that Monte Carlo risk analysis was developed. The statistical distributions with the 3 pt estimates are an acknowledgement that all estimates have a degree of uncertainty and the range is a model of how much or how little confidence people have in the original single point estimate.

In the example below, we can see how you might model a 10-day activity when you have either very high or very low confidence in the quality of the estimate. While they both have a 10 day base duration, the range between the low and the high estimates are a function of the quality of the estimate. In the image, the white bar represents the base duration (10 days) the blue bar represents the possible range of outcomes based on the quality of the estimate.